Time to Clinical Stability in Children With Community-Acquired Pneumonia.

BACKGROUND AND OBJECTIVES: Time to clinical stability (TCS) is a commonly used outcome in adults with community-acquired pneumonia (CAP), yet few studies have evaluated TCS in children. Our objective was to determine the association between TCS and disease severity in children with suspected CAP, as well as factors associated with reaching early stability. METHODS: This is a prospective cohort study of children (aged 3 months to 18 years) hospitalized with suspected CAP. TCS parameters included temperature, heart rate, respiratory rate, and hypoxemia with the use of supplemental oxygen. TCS was defined as time from admission to parameter normalization. The association of TCS with severity and clinical factors associated with earlier TCS were evaluated. RESULTS: Of 571 children, 187 (32.7%) had at least 1 abnormal parameter at discharge, and none had ≥3 abnormal discharge parameters. A greater proportion of infants (90 [93%]) had all 4 parameters stable at discharge compared with 12- to 18-year-old youths (21 [49%]). The median TCS for each parameter was <24 hours. Younger age, absence of vomiting, diffusely decreased breath sounds, and normal capillary refill were associated with earlier TCS. Children who did not reach stability were not more likely to revisit after discharge. CONCLUSIONS: A TCS outcome consisting of physiologic variables may be useful for objectively assessing disease recovery and clinical readiness for discharge among children hospitalized with CAP. TCS may decrease length of stay if implemented to guide discharge decisions. Clinicians can consider factors associated with earlier TCS for management decisions.

Bias between capnometry and venous carbon dioxide during initial assessment of pediatric emergency department patients: A video-based study.

Objective: The bias of capnometry (ETCO2) and venous carbon dioxide (vpCO2) among pediatric emergency department (PED) patients triaged to critical care areas is unknown. We aimed to explore correlations and bias between ETCO2 and vpCO2¸and identify predictors of bias. Methods: This was an observational, video-based, retrospective study comparing ETCO2 and vpCO2. Pediatric patients with simultaneous ETCO2 and vpCO2 data were included. Our primary aim utilized linear regressions to determine correlations and Bland-Altman analysis to assess bias. Our secondary aim utilized multiple regression to identify clinical covariates contributing to bias. Covariates included age, respiratory rate, heart rate, mean arterial blood pressure, capnometry interface, PED diagnosis, and PED disposition. Results: A total of 200 PED patients with ETCO2 and vpCO2 data were included. The median (interquartile range [IQR]) ETCO2, vpCO2, and ΔCO2 in mmHg were 38 (32, 46), 49 (41, 61), and 11 (4, 20), respectively. ETCO2 (r = 0.76) and ΔCO2 (r = 0.71) were highly correlated with vpCO2. The mean bias between ETCO2 and vpCO2 was -14.1 mmHg (95% confidence interval [CI], -41.9 -13.7). The bias between ETCO2 and vpCO2 increased at higher values of each measure. ETCO2 sampling interface was the only independent predictor of vpCO2 in our multivariate analysis. Patients requiring bag-valve mask (BVM) ventilation had the highest median bias between ETCO2 and vpCO2 (29 mmHg, IQR 15, 37). Conclusion: ETCO2 and vpCO2 were highly correlated. However, bias increased at higher levels of both ETCO2 and vpCO2. Among PED patients, ETCO2's ability to approximate vpCO2 diminishes with worsening hypercarbic respiratory failure.

Assessing the Revised Safer Dx Instrument® in the understanding of ambulatory system design changes for type 1 diabetes and autism spectrum disorder in pediatrics.

OBJECTIVES: We sought within an ambulatory safety study to understand if the Revised Safer Dx instrument may be helpful in identification of diagnostic missed opportunities in care of children with type 1 diabetes (T1D) and autism spectrum disorder (ASD). METHODS: We reviewed two months of emergency department (ED) encounters for all patients at our tertiary care site with T1D and a sample of such encounters for patients with ASD over a 15-month period, and their pre-visit communication methods to better understand opportunities to improve diagnosis. We applied the Revised Safer Dx instrument to each diagnostic journey. We chose potentially preventable ED visits for hyperglycemia, diabetic ketoacidosis, and behavioral crises, and reviewed electronic health record data over the prior three months related to the illness that resulted in the ED visit. RESULTS: We identified 63 T1D and 27 ASD ED visits. Using the Revised Safer Dx instrument, we did not identify any potentially missed opportunities to improve diagnosis in T1D. We found two potential missed opportunities (Safer Dx overall score of 5) in ASD, related to potential for ambulatory medical management to be improved. Over this period, 40 % of T1D and 52 % of ASD patients used communication prior to the ED visit. CONCLUSIONS: Using the Revised Safer Dx instrument, we uncommonly identified missed opportunities to improve diagnosis in patients who presented to the ED with potentially preventable complications of their chronic diseases. Future researchers should consider prospectively collected data as well as development or adaptation of tools like the Safer Dx.

Impact of Adjunct Corticosteroid Therapy on Quality of Life for Children With Suspected Pneumonia.

OBJECTIVE: To determine the association between adjunct corticosteroid therapy and quality of life (QoL) outcomes in children with signs and symptoms of lower respiratory tract infection and clinical suspicion for community-acquired pneumonia (CAP) in the emergency department (ED). METHODS: Secondary analysis from a prospective cohort study of children aged 3 months to 18 years with signs and symptoms of LRTI and a chest radiograph for suspected CAP in the ED, excluding children with recent (within 14 days) systemic corticosteroid use. The primary exposure was receipt of corticosteroids during the ED visit. Outcomes were QoL measures and unplanned visits. Multivariable regression was used to evaluate the association between corticosteroid therapy and outcomes. RESULTS: Of 898 children, 162 (18%) received corticosteroids. Children who received corticosteroids were more frequently boys (62%), Black (45%), had history of asthma (58%), previous pneumonia (16%), presence of wheeze (74%), and more severe illness at presentation (6%). Ninety-six percent were treated for asthma as defined by report of asthma or receipt of ß-agonist in the ED. Receipt of corticosteroids was not associated with QoL measures: days of activity missed (adjusted incident rate ratio [aIRR], 0.84; 95% confidence interval [CI], 0.63-1.11) and days of work missed (aIRR, 0.88; 95% CI, 0.60-1.27). There was a statistically significant interaction between age (>2 years) and corticosteroids receipt; the patients had fewer days of activity missed (aIRR, 0.62; 95% CI, 0.46-0.83), with no effect on children 2 years or younger (aIRR, 0.83; 95% CI, 0.54-1.27). Corticosteroid treatment was not associated with unplanned visit (odds ratio, 1.37; 95% CI, 0.69-2.75). CONCLUSIONS: In this cohort of children with suspected CAP, receipt of corticosteroids was associated with asthma history and was not associated with missed days of activity or work, except in a subset of children aged older than 2 years.

Viral Detection Is Associated With Severe Disease in Children With Suspected Community-Acquired Pneumonia.

OBJECTIVE: To evaluate the role of virus detection on disease severity among children presenting to the emergency department (ED) with suspected community-acquired pneumonia (CAP). METHODS: We performed a single-center prospective study of children presenting to a pediatric ED with signs and symptoms of a lower respiratory tract infection and who had a chest radiograph performed for suspected CAP. We included patients who had virus testing, with results classified as negative for virus, human rhinovirus, respiratory syncytial virus (RSV), influenza, and other viruses. We evaluated the association between virus detection and disease severity using a 4-tiered measure of disease severity based on clinical outcomes, ranging from mild ( discharged from the ED) to severe (receipt of positive-pressure ventilation, vasopressors, thoracostomy tube placement, or extracorporeal membrane oxygenation, intensive care unit admission, diagnosis of severe sepsis or septic shock, or death) in models adjusted for age, procalcitonin, C-reactive protein, radiologist interpretation of the chest radiograph, presence of wheeze, fever, and provision of antibiotics. RESULTS: Five hundred seventy-three patients were enrolled in the parent study, of whom viruses were detected in 344 (60%), including 159 (28%) human rhinovirus, 114 (20%) RSV, and 34 (6%) with influenza. In multivariable models, viral infections were associated with increasing disease severity, with the greatest effect noted with RSV (adjusted odds ratio [aOR], 2.50; 95% confidence interval [CI], 1.30-4.81) followed by rhinovirus (aOR, 2.18; 95% CI, 1.27-3.76). Viral detection was not associated with increased severity among patients with radiographic pneumonia (n = 223; OR, 1.82; 95% CI, 0.87-3.87) but was associated with severity among patients without radiographic pneumonia (n = 141; OR, 2.51; 95% CI, 1.40-4.59). CONCLUSIONS: The detection of a virus in the nasopharynx was associated with more severe disease compared with no virus; this finding persisted after adjustment for age, biomarkers, and radiographic findings. Viral testing may assist with risk stratification of patients with lower respiratory tract infections.

Incorporation of biomarkers into a prediction model for paediatric radiographic pneumonia.

Objective: The aim of this study was to evaluate biomarkers to predict radiographic pneumonia among children with suspected lower respiratory tract infections (LRTI). Methods: We performed a single-centre prospective cohort study of children 3 months to 18 years evaluated in the emergency department with signs and symptoms of LRTI. We evaluated the incorporation of four biomarkers (white blood cell count, absolute neutrophil count, C-reactive protein (CRP) and procalcitonin), in isolation and in combination, with a previously developed clinical model (which included focal decreased breath sounds, age and fever duration) for an outcome of radiographic pneumonia using multivariable logistic regression. We evaluated the improvement in performance of each model with the concordance (c-) index. Results: Of 580 included children, 213 (36.7%) had radiographic pneumonia. In multivariable analysis, all biomarkers were statistically associated with radiographic pneumonia, with CRP having the greatest adjusted odds ratio of 1.79 (95% CI 1.47-2.18). As an isolated predictor, CRP at a cut-off of 3.72 mg·dL-1 demonstrated a sensitivity of 60% and a specificity of 75%. The model incorporating CRP demonstrated improved sensitivity (70.0% versus 57.7%) and similar specificity (85.3% versus 88.3%) compared to the clinical model when using a statistically derived cutpoint. In addition, the multivariable CRP model demonstrated the greatest improvement in concordance index (0.780 to 0.812) compared with a model including only clinical variables. Conclusion: A model consisting of three clinical variables and CRP demonstrated improved performance for the identification of paediatric radiographic pneumonia compared with a model with clinical variables alone.

Developing Consensus on Clinical Outcomes for Children with Mild Pneumonia: A Delphi Study.

BACKGROUND: The absence of consensus for outcomes in pediatric antibiotic trials is a major barrier to research harmonization and clinical translation. We sought to develop expert consensus on study outcomes for clinical trials of children with mild community-acquired pneumonia (CAP). METHODS: Applying the Delphi method, a multispecialty expert panel ranked the importance of various components of clinical response and treatment failure outcomes in children with mild CAP for use in research. During Round 1, panelists suggested additional outcomes in open-ended responses that were added to subsequent rounds of consensus building. For Rounds 2 and 3, panelists were provided their own prior responses and summary statistics for each item in the previous round. The consensus was defined by >70% agreement. RESULTS: The expert panel determined that response to and failure of treatment should be addressed at a median of 3 days after initiation. Complete or substantial improvement in fever, work of breathing, dyspnea, tachypnea when afebrile, oral intake, and activity should be included as components of adequate clinical response outcomes. Clinical signs and symptoms including persistent or worsening fever, work of breathing, and reduced oral intake should be included in treatment failure outcomes. Interventions including receipt of parenteral fluids, supplemental oxygen, need for high-flow nasal cannula oxygen therapy, and change in prescription of antibiotics should also be considered in treatment failure outcomes. CONCLUSIONS: Clinical response and treatment failure outcomes determined by the consensus of this multidisciplinary expert panel can be used for pediatric CAP studies to provide objective data translatable to clinical practice.

Pediatric Emergency Medicine Physicians' Perceptions of Colleagues' Clinical Performance Over Career Span.

OBJECTIVE: The US physician workforce is aging, prompting concerns regarding clinical performance of senior physicians. Pediatric emergency medicine (PEM) is a high-acuity, multitasking, diagnostically complex and procedurally demanding specialty. Aging's impact on clinical performance in PEM has not been examined. We aimed to assess PEM physician's' perceptions of peers' clinical performance over career span. METHODS: We surveyed 478 PEM physician members of the American Academy of Pediatrics' Section on Emergency Medicine survey study list-serve in 2020. The survey was designed by the investigators with iterative input from colleagues. Respondents rated, using a 5-point Likert scale, the average performance of 4 age categories of PEM physicians in 9 clinical competencies. Additional items included concerns about colleague's performance and preferences for age of physician managing a critically ill child family member. RESULTS: We received 232 surveys with responses to core initial items (adjusted response rate, 49%). Most respondents were 36 to 49 (34.9%) or 50 to 64 (47.0%) years old. Fifty-three percent reported ever having concern about a colleague's performance. For critical care-related competencies, fewer respondents rated the ≥65-year age group as very good or excellent compared with midcareer physicians (36-49 or 50-64 years old). The ratings for difficult communications with families were better for those 65 years or older than those 35 years or younger. Among 129 of 224 respondents (58%) indicating a preferred age category for a colleague managing a critically ill child relative, most (69%) preferred a 36 to 49-year-old colleague. CONCLUSIONS: Pediatric emergency medicine physicians' perceptions of peers' clinical performance demonstrated differences by peer age group. Physicians 65 years or older were perceived to perform less well than those 36 to 64 years old in procedural and multitasking skills. However, senior physicians were perceived as performing as well if not better than younger peers in communication skills. Further study of age-related PEM clinical performance with objective measures is warranted.

Safer Type 1 Diabetes Care at Home: SEIPS-based Process Mapping with Parents and Clinicians.

Introduction: The limited data indicate that pediatric medical errors in the outpatient setting, including at home, are common. This study is the first step of our Ambulatory Pediatric Patient Safety Learning Lab to address medication errors and treatment delays among children with T1D in the outpatient setting. We aimed to identify failures and potential solutions associated with medication errors and treatment delays among outpatient children with T1D. Methods: A transdisciplinary team of parents, safety researchers, and clinicians used Systems Engineering Initiative for Patient Safety (SEIPS) based process mapping of data we collected through in-home medication review, observation of administration, chart reviews, parent surveys, and failure modes and effects analysis (FMEA). Results: Eight (57%) of the 14 children who had home visits experienced 18 errors (31 per 100 medications). Four errors in two children resulted in harm, and 13 had the potential for harm. Two injuries occurred when parents failed to treat severe hypoglycemia and lethargy, and two were due to repeated failures to administer insulin at home properly. In SEIPS-based process maps, high-risk errors occurred during communication between the clinic and home or in management at home. Two FMEAs identified interventions to better communicate with families and support home care, especially during evolving illness. Conclusion: Using SEIPS-based process maps informed by multimodal methods to identify medication errors and treatment delays, we found errors were common. Better support for managing acute illness at home and improved communication between the clinic and home are potentially high-yield interventions.

Antibiotic use and outcomes among children hospitalized with suspected pneumonia.

BACKGROUND: Although viral etiologies predominate, antibiotics are frequently prescribed for community-acquired pneumonia (CAP). OBJECTIVE: We evaluated the association between antibiotic use and outcomes among children hospitalized with suspected CAP. DESIGNS, SETTINGS AND PARTICIPANTS: We performed a secondary analysis of a prospective cohort of children hospitalized with suspected CAP. INTERVENTION: The exposure was the receipt of antibiotics in the emergency department (ED). MAIN OUTCOME AND MEASURES: Clinical outcomes included length of stay (LOS), care escalation, postdischarge treatment failure, 30-day ED revisit, and quality-of-life (QoL) measures from a follow-up survey 7-15 days post discharge. To minimize confounding by indication (e.g., radiographic CAP), we performed inverse probability treatment weighting with propensity analyses. RESULTS: Among 523 children, 66% were <5 years, 88% were febrile, 55% had radiographic CAP, and 55% received ED antibiotics. The median LOS was 41 h (IQR: 25, 54). After propensity analyses, there were no differences in LOS, escalated care, treatment failure, or revisits between children who received antibiotics and those who did not. Seventy-one percent of patients completed follow-up surveys after discharge. Among 16% of patients with fevers after discharge, the median fever duration was 2 days, and those who received antibiotics had a 37% decrease in the mean number of days with fever (95% confidence interval: 20% and 51%). We found no statistical differences in other QoL measures.

Serious Bacterial Infections in Young Febrile Infants With Positive Urinalysis Results.

It is unknown whether febrile infants 29 to 60 days old with positive urinalysis results require routine lumbar punctures for evaluation of bacterial meningitis. OBJECTIVE: To determine the prevalence of bacteremia and/or bacterial meningitis in febrile infants ≤60 days of age with positive urinalysis (UA) results. METHODS: Secondary analysis of a prospective observational study of noncritical febrile infants ≤60 days between 2011 and 2019 conducted in the Pediatric Emergency Care Applied Research Network emergency departments. Participants had temperatures ≥38°C and were evaluated with blood cultures and had UAs available for analysis. We report the prevalence of bacteremia and bacterial meningitis in those with and without positive UA results. RESULTS: Among 7180 infants, 1090 (15.2%) had positive UA results. The risk of bacteremia was higher in those with positive versus negative UA results (63/1090 [5.8%] vs 69/6090 [1.1%], difference 4.7% [3.3% to 6.1%]). There was no difference in the prevalence of bacterial meningitis in infants ≤28 days of age with positive versus negative UA results (∼1% in both groups). However, among 697 infants aged 29 to 60 days with positive UA results, there were no cases of bacterial meningitis in comparison to 9 of 4153 with negative UA results (0.2%, difference -0.2% [-0.4% to -0.1%]). In addition, there were no cases of bacteremia and/or bacterial meningitis in the 148 infants ≤60 days of age with positive UA results who had the Pediatric Emergency Care Applied Research Network low-risk blood thresholds of absolute neutrophil count <4 × 103 cells/mm3 and procalcitonin <0.5 ng/mL. CONCLUSIONS: Among noncritical febrile infants ≤60 days of age with positive UA results, there were no cases of bacterial meningitis in those aged 29 to 60 days and no cases of bacteremia and/or bacterial meningitis in any low-risk infants based on low-risk blood thresholds in both months of life. These findings can guide lumbar puncture use and other clinical decision making.

Feasibility and reliability of telemedicine examinations for respiratory distress in children: A pilot study.

Respiratory disorders are a leading cause of acute care visits by children. Data establishing the reliability of telemedicine in evaluating children with respiratory concerns are limited. The overall objective of this pilot study was to evaluate the use of telemedicine to evaluate children with respiratory concerns. We performed a pilot prospective cohort study of children 12 to 71 months old presenting to the emergency department (ED) with lower respiratory tract signs and symptoms. Three examinations were performed simultaneously-one by the ED clinician with the patient, one by a remote ED clinician using telemedicine, and one by the child's parent. We evaluated measures of agreement between (a) the local and remote clinicians, (b) the local clinician and the parent, and (c) the parent and the remote clinician. Twenty-eight patients were enrolled (84 paired examinations). Except for heart rate, all examination findings evaluated (general appearance, capillary refill time, grunting, nasal flaring, shortness of breath, retractions, impression of respiratory distress, respiratory rate, and temperature) had acceptable or excellent agreement between raters. In this pilot study, we found that telemedicine respiratory examinations of young children are feasible and reliable, using readily available platforms and equipment.

Distinct Characteristics and Chronology of Amoxicillin-Associated Reactions in Pediatric Acute Care Settings.

BACKGROUND: Amoxicillin-associated reactions (AARs) in children presenting as rashes are common, and recent data suggest that >90% tolerate amoxicillin on re-exposure. However, additional data would help pediatricians and allergists gain confidence in referring and testing children who experienced systemic symptoms perceived as "worrisome," thus leading to urgent medical evaluations. By characterizing the entire spectrum of AAR symptoms in pediatric patients presenting to emergency department (ED)/urgent care (UC) settings, we sought to increase our diagnostic acumen to guide subsequent allergy evaluations. OBJECTIVE: To fully characterize clinical features of rash and systemic symptoms in children presenting to the ED/UC with AARs. METHODS: A retrospective chart review of children seen in the ED/UC from July 1, 2015, to June 30, 2017, was conducted. Clinical features, chronology, and seasonality were detailed, and cases were classified into 3 previously described AAR phenotypes: maculopapular exanthem (MPE), urticaria, and serum sickness-like reactions (SSLRs), if they experienced joint symptoms. RESULTS: Children (n = 668; median age: 1.8 years) presented to the ED/UC with urticaria (44%), MPE (36%), and SSLRs (11%) typically on days 7 to 10 of amoxicillin. Although children with SSLRs were more frequently treated with corticosteroids (28%, P < .0001) and exhibited higher rates of "worrisome" features (fever, angioedema, or gastrointestinal symptoms; 73%, P < .0001), delayed-onset systemic symptoms were identified frequently in all 3 groups. ED/UC reutilization was unexpectedly high with 66 children (10%) returning to the ED/UC for re-evaluation. CONCLUSION: "Worrisome" symptoms are common in children presenting to the ED/UC with AARs. Future studies are needed to determine the impact on subsequent referral and allergy testing.

Predictors of Acute Care Reutilization in Pediatric Patients With Amoxicillin-Associated Reactions.

BACKGROUND: Amoxicillin-associated reactions (AARs) contribute to substantial health care utilization, with a reutilization rate of 10% in pediatric emergency department (ED) and urgent care (UC) settings. OBJECTIVE: To identify predictors of ED/UC reutilization by examining patients' clinical features and providers' management of AARs. METHODS: Through a retrospective chart review of 668 patients presenting with AARs over 2 years to the pediatric ED/UC, we examined clinical features associated with ED/UC reutilization, including rash phenotype, systemic symptoms (fever, angioedema, joint involvement, gastrointestinal symptoms), and providers' management (pharmacologic treatment and counseling). We then constructed a statistical model to predict ED/UC reutilization using stepwise backward model selection. RESULTS: ED/UC reutilizers were more likely to be male (P = .008) and have fever (P = .0001), angioedema (P < .0001), joint involvement (P < .0001), and gastrointestinal symptoms (P = .0001) during their AAR course. Rash phenotypes differed between groups (P < .0001), as ED/UC reutilizers more frequently exhibited urticaria. However, there were no differences in clinical management between groups, including pharmacologic recommendations, at the initial ED/UC encounter. In addition, our statistical model identified younger patients <2 years of age as more likely to reutilize ED/UC resources if providers did not document specific return precautions (odds ratio, 3.6; 95% confidence interval, 1.7-7.7). CONCLUSION: Recognition of clinical features and treatment gaps associated with ED/UC reutilization will guide interventions to optimize care in children presenting with AARs, such as improved anticipatory guidance and early allergy consultation. Prospective studies are needed to determine whether these interventions will reduce ED/UC reutilization and facilitate timely allergy testing.

Factors Associated With Antibiotic Use for Children Hospitalized With Pneumonia.

BACKGROUND: Antibiotics are frequently used for community-acquired pneumonia (CAP), although viral etiologies predominate. We sought to determine factors associated with antibiotic use among children hospitalized with suspected CAP. METHODS: We conducted a prospective cohort study of children who presented to the emergency department (ED) and were hospitalized for suspected CAP. We estimated risk factors associated with receipt of ≥1 dose of inpatient antibiotics and a full treatment course using multivariable Poisson regression with an interaction term between chest radiograph (CXR) findings and ED antibiotic use. We performed a subgroup analysis of children with nonradiographic CAP. RESULTS: Among 477 children, 60% received inpatient antibiotics and 53% received a full course. Factors associated with inpatient antibiotics included antibiotic receipt in the ED (relative risk 4.33 [95% confidence interval, 2.63-7.13]), fever (1.66 [1.22-2.27]), and use of supplemental oxygen (1.29 [1.11-1.50]). Children with radiographic CAP and equivocal CXRs had an increased risk of inpatient antibiotics compared with those with normal CXRs, but the increased risk was modest when antibiotics were given in the ED. Factors associated with a full course were similar. Among patients with nonradiographic CAP, 29% received inpatient antibiotics, 21% received a full course, and ED antibiotics increased the risk of inpatient antibiotics. CONCLUSIONS: Inpatient antibiotic utilization was associated with ED antibiotic decisions, CXR findings, and clinical factors. Nearly one-third of children with nonradiographic CAP received antibiotics, highlighting the need to reduce likely overuse. Antibiotic decisions in the ED were strongly associated with decisions in the inpatient setting, representing a modifiable target for future interventions.

Validation of Prediction Models for Pneumonia Among Children in the Emergency Department.

BACKGROUND: Several prediction models have been reported to identify patients with radiographic pneumonia, but none have been validated or broadly implemented into practice. We evaluated 5 prediction models for radiographic pneumonia in children. METHODS: We evaluated 5 previously published prediction models for radiographic pneumonia (Neuman, Oostenbrink, Lynch, Mahabee-Gittens, and Lipsett) using data from a single-center prospective study of patients 3 months to 18 years with signs of lower respiratory tract infection. Our outcome was radiographic pneumonia. We compared each model's area under the receiver operating characteristic curve (AUROC) and evaluated their diagnostic accuracy at statistically-derived cutpoints. RESULTS: Radiographic pneumonia was identified in 253 (22.2%) of 1142 patients. When using model coefficients derived from the study dataset, AUROC ranged from 0.58 (95% confidence interval, 0.52-0.64) to 0.79 (95% confidence interval, 0.75-0.82). When using coefficients derived from original study models, 2 studies demonstrated an AUROC >0.70 (Neuman and Lipsett); this increased to 3 after deriving regression coefficients from the study cohort (Neuman, Lipsett, and Oostenbrink). Two models required historical and clinical data (Neuman and Lipsett), and the third additionally required C-reactive protein (Oostenbrink). At a statistically derived cutpoint of predicted risk from each model, sensitivity ranged from 51.2% to 70.4%, specificity 49.9% to 87.5%, positive predictive value 16.1% to 54.4%, and negative predictive value 83.9% to 90.7%. CONCLUSIONS: Prediction models for radiographic pneumonia had varying performance. The 3 models with higher performance may facilitate clinical management by predicting the risk of radiographic pneumonia among children with lower respiratory tract infection.

Clinical Characteristics and Etiology of Community-acquired Pneumonia in US Children, 2015-2018.

BACKGROUND: Pneumonia has a major impact on childhood health and health care costs. This study was designed to obtain contemporary information on the clinical characteristics and etiology of community-acquired pneumonia (CAP) in children from both inpatient and outpatient settings in the USA. METHODS: We conducted a prospective, multicenter, observational study of CAP among previously healthy children 2 months to 18 years of age in 6 children's hospitals in Ohio from 2015 to 2018. For pathogen detection, nasopharyngeal swabs were collected from all subjects. Blood and pleural fluid cultures were available per standard of care. RESULTS: We enrolled a convenience sample of 441 patients: 380 hospitalized and 61 outpatients. Tachypnea and radiologic findings of consolidation and pleural effusion were more frequent among inpatients than outpatients. A pathogen was detected in 64.6% of patients: viruses in 55.6%, atypical bacteria in 8.8% and pyogenic bacteria in 4.3%. Eighteen (4.1%) patients had both viruses and bacteria detected. Rhinovirus/enterovirus (RV; 18.6%) and respiratory syncytial virus (RSV; 16.8%) were the viruses most frequently detected, and Mycoplasma pneumoniae (8.2%) and Streptococcus pneumoniae (2.3%) were the most common bacteria. Except for S. pneumoniae, which was identified more frequently in inpatients, there were no significant differences between inpatients and outpatients in the proportions of children with specific pathogens detected. CONCLUSIONS: Rhinovirus/enterovirus and RSV among viruses and M. pneumoniae and S. pneumoniae among bacteria were the most common pathogens detected in children with CAP. Tachypnea and chest radiographs with consolidation and/or pleural effusion were associated with hospitalization.

A Prediction Model for Pediatric Radiographic Pneumonia.

BACKGROUND: Chest radiographs (CXRs) are frequently used in the diagnosis of community-acquired pneumonia (CAP). We sought to construct a predictive model for radiographic CAP based on clinical features to decrease CXR use. METHODS: We performed a single-center prospective study of patients 3 months to 18 years of age with signs of lower respiratory infection who received a CXR for suspicion of CAP. We used penalized multivariable logistic regression to develop a full model and bootstrapped backward selection models to develop a parsimonious reduced model. We evaluated model performance at different thresholds of predicted risk. RESULTS: Radiographic CAP was identified in 253 (22.2%) of 1142 patients. In multivariable analysis, increasing age, prolonged fever duration, tachypnea, and focal decreased breath sounds were positively associated with CAP. Rhinorrhea and wheezing were negatively associated with CAP. The bootstrapped reduced model retained 3 variables: age, fever duration, and decreased breath sounds. The area under the receiver operating characteristic for the reduced model was 0.80 (95% confidence interval: 0.77-0.84). Of 229 children with a predicted risk of <4%, 13 (5.7%) had radiographic CAP (sensitivity of 94.9% at a 4% risk threshold). Conversely, of 229 children with a predicted risk of >39%, 140 (61.1%) had CAP (specificity of 90% at a 39% risk threshold). CONCLUSIONS: A predictive model including age, fever duration, and decreased breath sounds has excellent discrimination for radiographic CAP. After external validation, this model may facilitate decisions around CXR or antibiotic use in CAP.

Urinary Proadrenomedullin and Disease Severity in Children With Suspected Community-acquired Pneumonia.

BACKGROUND: Plasma proadrenomedullin (proADM) is a promising biomarker to predict disease severity in community-acquired pneumonia (CAP). Urinary biomarkers offer advantages over blood, including ease of collection. We evaluated the association between urinary proADM and disease severity in pediatric CAP. METHODS: We performed a prospective cohort study of children 3 months to 18 years with CAP. Urinary proADM/creatinine (Cr) was calculated. Disease severity was defined as: mild (discharged home), mild-moderate (hospitalized but not moderate-severe or severe), moderate-severe (eg, hospitalized with supplemental oxygen and complicated pneumonia) and severe (eg, vasopressors and invasive ventilation). Outcomes were examined using logistic regression within the cohort with suspected CAP and in a subset with radiographic CAP. RESULTS: Of the 427 children included, higher proADM/Cr was associated with increased odds of severe disease compared with nonsevere disease [suspected CAP, odds ratio (OR) 1.02 (95% confidence interval (CI) 1.003, 1.04); radiographic CAP, OR 1.03 (95% CI 1.01, 1.06)] when adjusted for other covariates. ProADM/Cr had an area under the receiver operating characteristic curve of 0.56 (threshold 0.9 pmol/mg) to differentiate severe from nonsevere disease in suspected CAP and 0.65 in radiographic CAP (threshold 0.82 pmol/mg). Healthy controls had less proADM in their urine (median, 0.61 pmol/mg) compared with suspected (0.87 pmol/mg, P = 0.018) and radiographic (0.73 pmol/mg, P = 0.016) CAP. CONCLUSIONS: Urinary proADM/Cr ratio measured at the time of emergency department visit was statistically associated with the development of severe outcomes in children with CAP, with stronger discriminatory performance in radiographic disease.

Respiratory Colonization and Short-Term Temporal Changes in the Urinary Metabolome of Children.

The human metabolome may vary based on age, over time, and in the presence of viral carriage and bacterial colonization-a common scenario in children. We used nuclear magnetic resonance spectroscopy to identify and quantify urinary metabolites of children without signs or symptoms of respiratory illness. A urine sample and two nasopharyngeal swabs were collected to test for respiratory viral pathogens and colonization by Streptococcus pneumoniae (Sp). Urine samples were collected at the initial visit, 24 h post-enrollment, and 10-14 days post-enrollment. Of the 122 children enrolled, 24% had a virus detected and 19.7% had Sp detected. Intraclass correlation coefficients demonstrated greater within-subject versus between-subject variability for all metabolites detected. In linear mixed models adjusted for age, time, history of asthma, Sp, and viruses, 1-methylnicotinamide was increased by 50% in children with Sp and decreased by 35% in children with rhinovirus/enterovirus. Children with Sp had 83% higher levels of trimethylamine-N-oxide compared with those without Sp. However, when adjusting for multiple comparisons, the association was no longer statistically significant. In conclusion, there appear to be short-term changes within the urinary metabolome of healthy children, but levels of metabolites did not statistically differ in children with viral carriage or Sp detected.